1. Field of the Invention
The present invention relates to electronic component substrates with integrated resilient spring contacts in the field of semiconductor devices, and more particularly to dice, wafers, Leadless Grid Array (“LGA”) sockets, and test head assemblies with resilient microelectronic spring contacts.
2. Description of Related Art
As semiconductor devices are made in increasingly smaller sizes, while at the same time becoming increasingly complex, semiconductor die size, and the size of contacts pads on such dice, has also shrunk. This trend towards ever-smaller and more powerful devices is projected to continue. The electronic components which use semiconductor devices are also being made in increasingly smaller sizes. New packaging technology, such as use of Chip Scale Packages (“CSPs”) has evolved in response to these trends towards smaller packages, and more dense arrays of contacts.
The trend towards use of CSPs has led to new requirements in the field of semiconductor manufacturing and component assembly. Nearly all present and proposed CSPs use a solder ball as the first level interconnect element. In the field of testing, such CSPs require a wafer or device-level contactor that can consistently and reliably make contact with solder balls without requiring a costly or time-consuming cleaning step after each use. The contactor should also require a low contact force, deliver low electrical resistance and parasitics, and survive numerous testing cycles (such as several thousand compression cycles at high temperature). The contactor should also scale easily, regardless of the number of dice per wafer, the number of terminals (contact pads or solder balls) per die, wafer diameter, terminal pitch, and electrical performance required. For example, current contactors should be capable of contacting as many as 100,000 terminals per wafer, at operating frequencies as high as one Gigahertz. Still higher densities and operating frequencies are anticipated in the future. Of course, the contactor must deliver all of this performance at an economically favorable cost.
Compact solder ball interconnect elements also place demanding requirements on assembly of CSPs onto Printed Circuit Boards (“PCBs”). Silicon, as used in CSPs, has a rate of thermal expansion about five times less than the material typically used in PCBS. A soldered joint between such mismatched materials is subject to stresses from thermal cycling, which over time can weaken the joint and degrade the electrical performance of the soldered CSP/PCB system. Traditional approaches, such as underfilling, can reduce problems caused by mismatched thermal properties, but such approaches are difficult to scale down to increasingly smaller sizes. In addition, the use of solder as a joining material creates a potential source of Alpha particles, which can reduce the reliability of adjacent semiconductor devices.
Microelectronic spring contacts made from relatively soft wire that is ball-bonded to terminals of a semiconductor device or contactor, then plated with a harder material for resiliency, have been used successfully with solder-ball type CSPs in the field of semiconductor testing. Exemplary spring contacts of this type, referred to herein as “composite contacts,” are disclosed, for example, in U.S. Pat. No. 5,476,211 (Khandros), which is incorporated herein by reference. Composite contacts have proven to be reliable and scaleable as required for modern semiconductor devices, and capable of repeatedly connecting to solder balls. Accordingly, composite contacts are well accepted in the field of semiconductor testing, where they are used on probe cards, interposers, Leadless Grid Arrays (“LGA”) sockets, and other such test substrates. Use of composite contacts on a wafer-level tester, including directly on a semiconductor wafer under test, is disclosed in U.S. Pat. No. 6,064,213 (Khandros et al.), which is incorporated herein, in its entirety, by reference. Attaching the spring contacts to the wafer or device under test (as opposed to a test substrate) offers certain advantages. These advantages include lower duty cycle requirements for the contact, and primarily, the opportunity to use the spring contact as the primary interconnection element during both testing and final assembly, thereby eliminating the need for solder balls.
However, each composite contact must be individually attached at its base by a wire bond. The economics of individual wire-bonding can become unfavorable at volume mass-production levels, such as when individually attaching spring contacts to tens of thousands of terminals on a wafer containing high-volume production semiconductor devices. Hence, use of composite contacts has generally been limited to test substrates, such as probe card assemblies and LGA production sockets, or to relatively low-volume, high-performance devices. It has not yet been possible to provide improved resilient contact elements with performance as good or better than composite contacts, but that can also be mass-produced at a lower cost. It is desired, therefore, to provide wafer and semiconductor devices with such improved resilient contact elements. It is further desired to provide electronic component substrates, such as probe cards, wafer contactors, LGA sockets, and test head assemblies, with the benefits of such improved resilient contact elements.